1. Region of the Invention
The present invention relates to a device and a method capable of automatically determining a color of an image.
2. Description of the Related Art
A known image forming device reproduces an original image onto a printing medium such as paper. A color copy machine is one example of such an image forming device. The color copy machine can reproduce a scanned original image by full color. However, when the original image is expressed by a single-color color, it is inefficient for the color copy machine to operate engines of four colors, Yellow (Y), Magenta (M), Cyan (C) and black (K). If an image forming process can be carried out by using only a toner corresponding to a color component included in the original image, the image forming process can be carried out efficiently.
For such a purpose, the color of the original image is required to be determined. When a user carries out a color determination visually, a wrong determination may be made due to the subjectivity of the user. Even when the user determines the color correctly, the user is required to designate a hue by a key input operation or the like. Since such a key input operation is required, an inefficiency remains.
Due to such a circumstance, there exists a demand for an image forming device which includes a device for automatically determining a color of an original image.
Regarding an automatic determination of a hue, a known device includes a color identification circuit which determines the hue for a purpose of separating a color of a format part when an original image is read by an Optical Character Reader (OCR). The color identification circuit converts an image signal expressed by three color components of Red (R), Green (G) and Blue (B) into a color space of HSV (H=hue, S=saturation, V=luminance). Then, the color identification circuit compares a prescribed threshold value with a signal level, and determines the hue.
A known color separating device can separate colors without using a fixed threshold value. In a color separating process carried out by the color separating device, a color difference for each pixel is obtained from an image signal of three color components of RGB for each pixel. Then, according to a moving average, the color separating device clusters closely related colors among each pixel.
However, the known devices do not consider sufficiently about how to set a boundary, which is to be a standard for determining the color.
That is, the color identification circuit arranges a plurality of straight lines as color determination boundary lines in a color space (plane) defined by two color difference variables. However, the straight lines are defined just under a condition to distinguish a hue. A consideration beyond this extent is not made.
Therefore, a determination is required to be carried out as to a determining color belongs to a region located in which side of a color determination boundary line. Such a determination is carried out by calculating in which position and in which direction in the two-dimensional color space (color plane) a coordinate point of the determining color is located, and comparing a positional relation between the calculated position and the color determination boundary line. In such a calculation, a calculation for calculating in which direction from an origin a position to be determined is located becomes a complicated calculation since the calculation accompanies a calculation of an irrational number such as an inverse trigonometric function. As a result, the color determination is inefficient.
The above-described color separating device can carry out a color separation by using a moving average. At a stage when specifically determining the hue of the color separated as described above, the separated color is compared with a threshold value as a hue boundary. However, a particular improvement has not been made as to a selection and a comparison of the threshold value.
The above-described problems occur in the color copy machine and also in other machines such as a color facsimile machine. For example, in the color facsimile machine, the above-described problems occur when automatically determining by which color a received image will be printed in a single-color printing operation.